Updated 1d ago
by nerdai
llm-agents-from-scratch
Build LLM agents and multi-agent systems from scratch, with MCP, Skills, and A2A
194 GitHub stars (194)
+5 in 30 days
| Period | New stars |
|---|---|
| Oct 1 | 0 |
| Oct 2 | +1 |
| Oct 3 | 0 |
| Oct 4 | 0 |
| Oct 5 | 0 |
| Oct 6 | +1 |
| Oct 7 | 0 |
| Oct 8 | 0 |
| Oct 9 | 0 |
| Oct 10 | 0 |
| Oct 11 | +1 |
| Oct 12 | 0 |
| Oct 13 | +2 |
| Oct 14 | 0 |
| Oct 15 | 0 |
| Oct 16 | 0 |
| Oct 17 | 0 |
| Oct 18 | 0 |
| Oct 19 | 0 |
| Oct 20 | 0 |
| Oct 21 | 0 |
| Oct 22 | 0 |
| Oct 23 | 0 |
| Oct 24 | 0 |
| Oct 25 | 0 |
| Oct 26 | 0 |
| Oct 27 | 0 |
| Oct 28 | 0 |
| Oct 29 | 0 |
| Oct 30 | 0 |
| Oct 31 | 0 |
| Nov 1 | +1 |
| Nov 2 | 0 |
| Nov 3 | 0 |
| Nov 4 | +1 |
| Nov 5 | 0 |
- 11 skills (11 skills found in its SKILL.md files)
- 69 forks (69 forks)
- 1d ago last update (Last updated 1d ago)
Install
Skills CLI
npx skills add nerdai/llm-agents-from-scratch
Asks which skills and agents to install. Just one: npx skills add nerdai/llm-agents-from-scratch --skill benchmark-ollama-models — add -g for every project.
Claude Code
git clone https://github.com/nerdai/llm-agents-from-scratchcp -r "llm-agents-from-scratch/.claude/skills/benchmark-ollama-models" ~/.claude/skills/
Copies the benchmark-ollama-models skill into ~/.claude/skills/ for every project (use .claude/skills/ for one project). Repeat for the other skills listed below.
Plugin
/plugin marketplace add nerdai/llm-agents-from-scratch
Run inside Claude Code if the repository ships a plugin marketplace, then pick plugins from /plugin.
Skills can run code on your machine — read the SKILL.md first.
What it does
In the author's own words, from each skill's SKILL.md.
-
benchmark-ollama-models
Benchmark Ollama Cloud models on the chapter notebooks' behaviour (real tool calls, skill activation, subagents finishing), not just pass/fail, to pick or re-validate the default Ollama Cloud model. Use before changing make_llm()'s Ollama Cloud default, when a cloud model is retired, or when asked to compare or test cloud models on the notebooks.
.claude/skills/ on GitHub (opens in a new tab)benchmark-ollama-models/ SKILL.md -
create-issue
Create a GitHub issue in nerdai/llm-agents-from-scratch and add it to project
.claude/skills/ on GitHub (opens in a new tab)create-issue/ SKILL.md -
version-bump
Bump the package version (major, minor, or patch), then sync CITATION.cff (version + date-released), src/_version.py, and CHANGELOG.md. Use when asked to cut a release or bump the version.
.claude/skills/ on GitHub (opens in a new tab)version-bump/ SKILL.md -
data_structures
No description in the SKILL.md.
docs/api_reference/ on GitHub (opens in a new tab)data_structures/ skill.md -
skills
No description in the SKILL.md.
docs/api_reference/ on GitHub (opens in a new tab)skills/ skill.md -
stop-at-one
Compute a sequence for a given starting number by repeatedly calling the next_number tool until the value 1 is reached.
chapter-notebooks/shared/ on GitHub (opens in a new tab).agents/ skills/ stop-at-one/ SKILL.md -
word-frequency
Compute the top-10 most frequent words in any text passage supplied by the user.
more-examples/ch06/ on GitHub (opens in a new tab).agents/ skills/ word-frequency/ SKILL.md -
email-drafter
Draft a professional reply to an email provided by the user.
more-examples/ch08/ on GitHub (opens in a new tab).agents/ skills/ email-drafter/ SKILL.md -
meeting-notes-summarizer
Summarise a meeting transcript by first asking the human operator for their preferred format and level of detail, then producing the summary accordingly.
more-examples/ch08/ on GitHub (opens in a new tab).agents/ skills/ meeting-notes-summarizer/ SKILL.md -
factor-step-by-step
Fully factor a number into primes by repeatedly calling the factor_step tool, one prime factor at a time, until the remainder is 1.
more-examples/ch09/ on GitHub (opens in a new tab).agents/ skills/ factor-step-by-step/ SKILL.md -
hailstone-sequence
Compute the complete Hailstone sequence for a given starting number by repeatedly calling the next_number tool until the value 1 is reached.
talks/2026/ on GitHub (opens in a new tab)acm-educational-talks/ .agents/ skills/ hailstone-sequence/ SKILL.md
No skill matches that filter.
Review before you install
A skill is a set of instructions and scripts your agent will follow, and it can run code on your machine. Read the SKILL.md and any scripts first — this listing is not a security review.